ABSTRACT The successful application of adoptive cell transfer (ACT) using chimeric antigen receptor (CAR)-transduced T- cells (CAR-T-cells) in solid tumors, such as GBM, will require improved tumor-homing ability by the T-cells. Recent studies of autoimmune conditions in the central nervous system (CNS) show that interleukin-17 (IL-17)- producing Type17 T-cells, which are reactive to CNS autoantigens, have a higher migratory capability to the CNS parenchyma than other T-cell subpopulations due to their ability to penetrate the blood brain barrier (BBB) via expression of IL-17/IL-22. Moreover, a chemokine receptor, CCR6, on Type17 T-cells allows them to infiltrate the CNS by interacting with its ligand, CCL20. Although the induction of CNS autoimmunity should be avoided, these data led us to hypothesize that Type17 T-cells engineered to direct glioblastoma (GBM) antigens will safely demonstrate superior anti-glioma activity compared with Type1 T-cells. To evaluate our hypothesis in preclinical studies, we will employ both a syngeneic mouse glioma model (Aim 1) and a human xenograft model and ACT with human GBM patient-derived CAR-T-cells (Aim 2). Aim 1. Mouse Type17 T-cells will demonstrate superior in vivo persistence, tumor-homing, and anti- tumor effects without CNS autoimmunity in syngeneic glioma models compared with Type1 T-cells. In our preliminary data, Type17 T-cells showed a more sustained anti-tumor immune response in vitro than did Type1 counterparts. Furthermore, our Type17 T-cells express both CXCR3 and CCR6 and induce glioma cells to produce their cognate ligands, CXCL10 and CCL20, respectively. Based on these data, we hypothesize that Type17 T-cells will demonstrate superior therapeutic efficacy against intra-cerebral gliomas compared with Type1 cells due to their greater in vivo persistence and their ability to traffic to CNS tumors. Using a C57BL/6 mouse syngeneic model, we will determine whether the CXCL10-CXCR3 and CCL20-CCR6 chemokine axes promote Type17 cell infiltration of CNS gliomas. We will also ensure the absence of CNS autoimmunity. Aim 2. Human GBM patient-derived Type17 T-cells transduced with CAR will safely mediate anti-tumor effects against human GBM xenograft in vivo. We will evaluate whether we can propagate Type17 CAR-T cells using GBM patient-derived T-cells, and whether the critical phenotype (e.g., chemokine receptor expression) observed in the syngeneic model in Aim 1 can be recapitulated in human GBM patient-derived Type17 T-cells that are transduced with the CAR. Finally, using immunocompromised NOD/scid/c(-/-) (NSG) mice bearing human GBM patient-derived xenografts in the brain, we will determine whether intravenous (i.v.) administration of human Type17 CAR-T-cells, especially both EGFRvIII- and EphA2-CARs compared with the single antigen-targeting approach, achieves effective and long-lasting therapeutic response against the human GBM xenografts without causing significant CNS toxicity.